Generative adversarial networks are used to achieve generative modeling in Deep Learning. It is an unsupervised task that involves the discovery of patterns in the input data to generate the output.
The generator is used to generate new examples, while the discriminator is used to classify the examples generated by the generator.
